R. Paul Robertson

Chronic oxidative stress as a central mechanism for glucose toxicity in pancreatic islet beta cells in diabetes.

Glucose in chronic excess causes toxic effects on structure and function of organs, including the pancreatic islet. Multiple biochemical pathways and mechanisms of action for glucose toxicity have been suggested. These include glucose autoxidation, protein kinase C activation, methylglyoxal formation and glycation, hexosamine metabolism, sorbitol formation, and oxidative phosphorylation. There are many potential mechanisms whereby excess glucose metabolites traveling along these pathways might cause beta cell damage. However, all these pathways have in common the formation of reactive oxygen species that, in excess and over time, cause chronic oxidative stress, which in turn causes defective insulin gene expression and insulin secretion as well as increased apoptosis. This minireview provides an overview of these mechanisms, as well as a consideration of whether antioxidant strategies might be used to protect further deterioration of the beta cell after the onset of diabetes and hyperglycemia. Diabetes mellitus is a disease characterized by hyperglycemia and is caused by absolute or relative insulin deficiency, sometimes associated with insulin resistance. It has multiple etiologies and segregates into two major forms. Type 1 diabetes is an autoimmune disease in which the patient’s own immune system reacts against islet antigens and destroys the beta cell. Type 2 diabetes is a polygenic syndrome with multiple etiologies rather than a single specific disease. As the hyperglycemia of diabetes becomes chronic, the sugar that normally serves as substrate, fuel, and signal takes on the darker role of toxin. Chronic hyperglycemia is the proximate cause of retinopathy, kidney failure, neuropathies, and macrovascular disease in diabetes. The beta cell in type 2 diabetes is also adversely affected by chronic hyperglycemia and, in this sense, is also a target for secondary complications. As hyperglycemia worsens, the beta cell steadily undergoes deterioration, secretes less and less insulin, and becomes a participant in a downward spiral of loss of function. This relentless deterioration in cell function caused by constant exposure to supraphysiologic concentrations of glucose is termed glucose toxicity. Mechanisms of Hyperglycemia-induced Oxidative Stress In physiologic concentrations, endogenous reactive oxygen species (ROS) help to maintain homeostasis. However, when ROS accumulate in excess for prolonged periods of time, they cause chronic oxidative stress and adverse effects. This is particularly relevant and dangerous for the islet, which is among those tissues that have the lowest levels of intrinsic antioxidant defenses. Multiple biochemical pathways and mechanisms of action have been implicated in the deleterious effects of chronic hyperglycemia and oxidative stress on the function of vascular, retinal, and renal tissues. Considerably less work has been performed using islet tissue. At least six pathways are emphasized in the literature as being major contributors of ROS. Each will be considered briefly. Glyceraldehyde Autoxidation—Glyceraldehyde 3-phosphate is a phosphorylation product formed from glucose during anaerobic glycolysis. The partner product, dihydroxyacetone phosphate, also contributes to intracellular glyceraldehyde concentrations via enzymatic conversion by triose-phosphate isomerase. Thereafter, glyceraldehyde 3-phosphate is oxidized by glyceraldehyde-phosphate dehydrogenase (GAPDH). Continuance of glycolysis yields pyruvate, which enters the mitochondria where it is oxidized to acetyl-CoA, and the processes of the tricarboxylic acid cycle and oxidative phosphorylation begin. One alternative to this classic pathway of glucose metabolism is the less familiar route of glyceraldehyde autoxidation (Fig. 1, pathway 1). The potential relevance of this pathway to diabetes mellitus was pointed out by Wolff and Dean (1), who emphasized that autoxidation of -hydroxyaldehydes generates hydrogen peroxide (H2O2) and -ketoaldehydes. In the presence of redox active metals, H2O2 can form the highly toxic hydroxyl radical. This pathway, therefore, forms two potentially toxic substances, -ketoaldehydes, which contribute to glycosylation-related protein chromophore development, and the hydroxyl radical, a reactive oxygen species that can cause mutagenic alterations in DNA. Although glyceraldehyde is characteristically thought of as an insulin secretagogue, when present in excess it may also inhibit insulin secretion (2). Long term exposure to high glucose concentrations decreases GAPDH activity in islets (3), which favors excess glyceraldehyde accumulation. Exposure of endothelial cells to 30 mM glucose caused GAPDH inhibition (4) through the mechanism of ROS-activated poly(ADP-ribosyl)ation of GAPDH by poly(ADP-ribose) polymerase. This in turn was associated with intracellular advanced glycation end product (AGE) formation and activation of PKC, the hexosamine pathway, and NFB. PKC Activation—Dihydroxyacetone can undergo reduction to glycerol 3-phosphate and acylation and thereby increase de novo synthesis of diacylglycerol, which activates protein kinase C, of which there at least 11 isoforms (Fig. 1, pathway 2). Activation of PKC has many biochemical consequences that relate to microvascular disease in diabetes. PKC activation is associated with increases in TGF1, vascular endothelial

Minireview: Secondary β-Cell Failure in Type 2 Diabetes—A Convergence of Glucotoxicity and Lipotoxicity

Chronic hyperglycemia and hyperlipidemia can exert deleterious effects on -cell function, respectively referred to as glucotoxicity and lipotoxicity. Over time, both contribute to the progressive deterioration of glucose homeostasis characteristic of type 2 diabetes. The mechanisms of glucotoxicity involve several transcription factors and are, at least in part, mediated by generation of chronic oxidative stress. Lipotoxicity is probably mediated by accumulation of a cytosolic signal derived from the fatty acid esterification pathway. Our view that hyperglycemia is a prerequisite for lipotoxicity is supported by several recent studies performed in our laboratories. First, prolonged in vitro exposure of isolated islets to fatty acids decreases insulin gene expression in the presence of high glucose concentrations only, and glucose is ratelimiting for the incorporation of fatty acids into neutral lipids. Second, normalization of blood glucose in Zucker diabetic fatty rats prevents accumulation of triglycerides and impairment of insulin gene expression in islets, whereas normalization of plasma lipid levels is without effect. Third, high-fat feeding in Goto-Kakizaki rats significantly impairs glucoseinduced insulin secretion in vitro, whereas a similar diet has no effect in normoglycemic animals. We propose that chronic hyperglycemia, independent of hyperlipidemia, is toxic for -cell function, whereas chronic hyperlipidemia is deleterious only in the context of concomitant hyperglycemia. (Endocrinology 143: 339 –342, 2002)

How Do We Define Cure of Diabetes

The mission of the American Diabetes Association is “to prevent and cure diabetes and to improve the lives of all people affected by diabetes.” Increasingly, scientific and medical articles (1) and commentaries (2) about diabetes interventions use the terms “remission” and “cure” as possible outcomes. Several approved or experimental treatments for type 1 and type 2 diabetes (e.g., pancreas or islet transplants, immunomodulation, bariatric/metabolic surgery) are of curative intent or have been portrayed in the media as a possible cure. However, defining remission or cure of diabetes is not as straightforward as it may seem. Unlike “dichotomous” diseases such as many malignancies, diabetes is defined by hyperglycemia, which exists on a continuum and may be impacted over a short time frame by everyday treatment or events (medications, diet, activity, intercurrent illness). The distinction between successful treatment and cure is blurred in the case of diabetes. Presumably improved or normalized glycemia must be part of the definition of remission or cure. Glycemic measures below diagnostic cut points for diabetes can occur with ongoing medications (e.g., antihyperglycemic drugs, immunosuppressive medications after a transplant), major efforts at lifestyle change, a history of bariatric/metabolic surgery, or ongoing procedures (such as repeated replacements of endoluminal devices). Do we use the terms remission or cure for all patients with normal glycemic measures, regardless of how this is achieved? A consensus group comprised of experts in pediatric and adult endocrinology, diabetes education, transplantation, metabolism, bariatric/metabolic surgery, and (for another perspective) hematology-oncology met in June 2009 to discuss these issues. The group considered a wide variety of questions, including whether it is ever accurate to say that a chronic illness is cured; what the definitions of management, remission, or cure might be; whether goals of managing comorbid conditions revert to those of patients without diabetes if someone is …

Inactivation of specific β cell transcription factors in type 2 diabetes

Type 2 diabetes (T2DM) commonly arises from islet β cell failure and insulin resistance. Here, we examined the sensitivity of key islet-enriched transcription factors to oxidative stress, a condition associated with β cell dysfunction in both type 1 diabetes (T1DM) and T2DM. Hydrogen peroxide treatment of β cell lines induced cytoplasmic translocation of MAFA and NKX6.1. In parallel, the ability of nuclear PDX1 to bind endogenous target gene promoters was also dramatically reduced, whereas the activity of other key β cell transcriptional regulators was unaffected. MAFA levels were reduced, followed by a reduction in NKX6.1 upon development of hyperglycemia in db/db mice, a T2DM model. Transgenic expression of the glutathione peroxidase-1 antioxidant enzyme (GPX1) in db/db islet β cells restored nuclear MAFA, nuclear NKX6.1, and β cell function in vivo. Notably, the selective decrease in MAFA, NKX6.1, and PDX1 expression was found in human T2DM islets. MAFB, a MAFA-related transcription factor expressed in human β cells, was also severely compromised. We propose that MAFA, MAFB, NKX6.1, and PDX1 activity provides a gauge of islet β cell function, with loss of MAFA (and/or MAFB) representing an early indicator of β cell inactivity and the subsequent deficit of more impactful NKX6.1 (and/or PDX1) resulting in overt dysfunction associated with T2DM.

Differentiating Glucose Toxicity From Glucose Desensitization: A New Message From the Insulin Gene

Our perspective is that the concepts of glucose toxicity and glucose desensitization should be differentiated because they carry very different connotations. The term glucose desensitization most properly refers to a pharmacological event involving a temporary, readily induced, physiological and reversible state of cellular refractoriness because of repeated or prolonged exposure to high concentrations of glucose. The term glucose toxicity should be reserved for nonphysiological, irreversible alterations in cellular function caused by chronic exposure to high glucose concentrations. With regard to the pancreatic islet β-cell, the mechanism of action for glucose desensitization seems most likely to be expressed at the level of the insulin exocytotic apparatus or insulin stores within the β-cell, whereas the mechanism of action for glucose toxicity may be at the level of insulin gene transcription. This differentiation raises the possibility that exposure of patients to chronic hyperglycemia may cause glucose toxic effects on the process of insulin gene transcription and/or expression that are irreversible. If so, this may contribute to so-called secondary drug failure and, in any event, reemphasizes the need to intensify therapeutic efforts to better regulate glycemia in type II diabetes.

Chronic oxidative stress as a mechanism for glucose toxicity of the beta cell in Type 2 diabetes

Type 2 diabetes is characterized by a relentless decline in pancreatic islet beta cell function and worsening hyperglycemia despite optimal medical treatment. Our central hypothesis is that residual hyperglycemia, especially after meals, generates reactive oxygen species (ROS), which in turn causes chronic oxidative stress on the beta cell. This hypothesis is supported by several observations. Exposure of isolated islets to high glucose concentrations induces increases in intracellular peroxide levels. The beta cell has very low intrinsic levels of antioxidant proteins and activities and thus is very vulnerable to ROS. Treatment with antioxidants protects animal models of type 2 diabetes against complete development of phenotypic hyperglycemia. The molecular mechanisms responsible for the glucose toxic effect on beta cell function involves disappearance of two important regulators of insulin promoter activity, PDX-1 and MafA. Antioxidant treatment in vitro prevents disappearance of these two transcription factors and normalizes insulin gene expression. These observations suggest that the ancillary treatment with antioxidants may improve outcomes of standard therapy of type 2 diabetes in humans.

Effect of the two-layer (University of Wisconsin solution-perfluorochemical plus O2) method of pancreas preservation on human islet isolation, as assessed by the Edmonton Isolation Protocol.

Background. Current techniques for isolating islets require that pancreata stored with University of Wisconsin solution (UW) are processed within 12 hours of cold storage. In this study, we hypothesized that the two-layer method (TLM) could extend the acceptable preservation period of pancreata before islet isolation and increase islet yields. Methods. In the first experimental set, eight pancreata were maintained for an average of 8.3±1.2 hours in UW and transferred into the TLM for an additional 14.3±1.1 hours for a total cold ischemic period of 22.6±1.6 hours (prolonged TLM). Four pancreata were maintained as a control group in UW alone for a total of 21.3±2.0 hours. In the second experimental set, six pancreata were maintained for an average of 6.4±1.8 hours in UW followed by 4.8±0.8 hours with the TLM for a total preservation time of 11.3±2.5 hours (short TLM). The control organs for the short TLM group were stored for an average of 9.5±1.3 hours in UW alone. Islets were isolated and evaluated according to the Edmonton protocol. Results. Between each group of the two experimental sets, there was no significant difference in donor-related factors (i.e. gender, age, body mass index [BMI], etc.). The TLM as compared with UW preservation resulted in a significant increase in islet yields postpurification for both short (3,353±394 islet equivalents [IE] vs. 2,027±415 IE; mean±SEM) and prolonged (2,404±503 IE vs. 514±180 IE) periods of storage. Furthermore, islet yields after prolonged storage with the TLM were not significantly different from organs maintained for only a short period with UW (P =0.17). The quality of islets as assessed by size, postculture viability, survival rates, insulin content, and insulin secretion were similar for each of the four groups. Conclusion. In comparison with UW organ preservation, exposure of pancreata to the TLM result in greater islet yields and extended preservation times.

Pancreatic endocrine function in cystic fibrosis

To characterize pancreatic endocrine secretion and to examine interrelationships among alterations in alpha, beta, and pancreatic polypeptide cell function in patients with cystic fibrosis (CF), we studied 19 patients with exocrine insufficiency (EXO), including 9 receiving insulin therapy (EXO-IT); 10 patients with no exocrine insufficiency (NEXO); and 10 normal control subjects. First-phase C-peptide response to intravenously administered glucose was significantly impaired in CF patients with exocrine insufficiency (EXO-IT = 0.02 +/- 0.01; EXO = 0.11 +/- 0.02; NEXO = 0.25 +/- 0.05; control subjects = 0.30 +/- 0.04 nmol/L). Lowering fasting glucose levels with exogenous insulin administration in EXO-IT did not improve beta cell responsivity to glucose. The C-peptide response to arginine was less impaired (EXO-IT = 0.12 +/- 0.02; EXO = 0.15 +/- 0.02; NEXO = 0.23 +/- 0.06; control subjects = 0.28 +/- 0.04 nmol/L). Alpha cell function, measured as peak glucagon secretion in response to hypoglycemia, was diminished in EXO but not NEXO (EXO-IT = 21 +/- 10; EXO = 62 +/- 19; NEXO = 123 +/- 29; control subjects = 109 +/- 12 ng/L). Despite diminished glucagon response, EXO patients recovered normally from hypoglycemia. Peak pancreatic polypeptide response to hypoglycemia distinguished CF patients with exocrine insufficiency from those without exocrine insufficiency (EXO-IT = 3 +/- 2; EXO = 3 +/- 1; NEXO = 226 +/- 68; control subjects = 273 +/- 100 pmol/L). Thus CF patients with exocrine disease have less alpha, beta, and pancreatic polypeptide cell function than CF patients without exocrine disease. These data suggest either that exocrine disease causes endocrine dysfunction in CF or that a common pathogenic process simultaneously and independently impairs exocrine and endocrine function.

Preserved Insulin Secretion and Insulin Independence in Recipients of Islet Autografts

BACKGROUND Transplantation of pancreatic islets, rather than whole pancreas, has been introduced as a treatment for diabetes mellitus. We studied five patients ranging in age from 12 to 37 years who had severe chronic pancreatitis for which they underwent total pancreatectomy followed by isolation and hepatic transplantation of their own islets. METHODS All patients had remained insulin-independent for 1 to 7 1/2 years after transplantation. The numbers of islets transplanted ranged from 110,000 to 412,000. Islet function was assessed by measuring the plasma insulin responses to intravenous glucose and arginine and the plasma glucagon responses to hypoglycemia and arginine. In one patient, islet function was studied during catheterization of the hepatic vein, portal vein, and splenic artery and by analysis of a liver-biopsy specimen. RESULTS After transplantation, the mean (+/- SD) fasting plasma glucose concentration was 122 +/- 47 mg per deciliter (6.8 +/- 2.6 mmol per liter) and the hemoglobin A1c concentration was 6.0 +/- 0.8 percent in the five patients. The values were most abnormal--214 mg per deciliter (11.9 mmol per liter) and 7.3 percent, respectively--in the patient who received only 110,000 islets. The acute plasma insulin responses to glucose and to arginine in the five patients were 23 +/- 13 and 26 +/- 10 microU per milliliter (168 +/- 94 and 184 +/- 70 pmol per liter), respectively, as compared with 58 +/- 6 and 37 +/- 8 microU per milliliter (416 +/- 44 and 267 +/- 61 pmol per liter) in the normal subjects. The peak plasma glucagon responses to insulin and arginine were 21 +/- 4 and 65 +/- 36 pg per milliliter, respectively, as compared with 125 +/- 28 and 156 +/- 99 pg per milliliter in the normal subjects. All five patients had plasma epinephrine but not pancreatic polypeptide responses to hypoglycemia. The results of the hepatic-vein catheterization in one patient indicated that the transplanted islets released insulin and glucagon in response to arginine. Immunoperoxidase staining of this patients liver-biopsy specimen showed that the islets contained insulin, glucagon, and somatostatin but not pancreatic polypeptide. CONCLUSIONS Intrahepatic transplantation of as few as 265,000 islets can result in the release of insulin and glucagon at appropriate times and in prolonged periods of insulin independence.

A role for prostaglandin E in defective insulin secretion and carbohydrate intolerance in diabetes mellitus.

Prostaglandin E(2) (PGE(2)) infusion in normal humans inhibited acute insulin responses to a glucose (5 g i.v.) pulse (response before PGE(2) = 593 +/- 104%; during PGE(2) = 312+/-55%; mean+/-SE, mean change 3-5 min insulin,% basal, P < 0.005). This effect was associated with a decrease in glucose disappearance rates (K(G) before PGE(2) = 0.73+/-0.07; during PGE(2) = 0.49+/-0.06%/min, P < 0.025). Acute insulin responses to arginine (2 g i.v.) were not affected by PGE(2) (response before PGE(2) = 592+/-164%; during PGE(2) = 590+/-118%; P = NS). Infusion of sodium salicylate (SS), an inhibitor of endogenous prostaglandin synthesis, augmented acute insulin responses to glucose in normals (response before SS = 313+/-62%; during SS = 660+/-86%; P < 0.001). In adult-onset diabetes with fasting hyperglycemia, SS restored absent acute insulin responses to glucose (20 g i.v.) pulses (response before SS = 5+/-6%; during SS = 97+/-24%; P < 0.005). This was accompanied by a fourfold augmentation in second phase insulin secretion (second phase before SS = 1,696+/-430%; during SS = 5,176+/-682%; change 10-60 min insulin, muU/ml.min,% basal, P < 0.001) and by acceleration of glucose disappearance rates (K(G) before SS = 0.56+/-0.06; during SS = 1.02+/-0.17%/min, P < 0.005). These findings uniquely demonstrate that (a) PGE(2) inhibits glucose-induced acute insulin responses and decreases glucose disposal in nondiabetic humans and (b) SS restores acute insulin responses, augments second phase insulin secretion, and accelerates glucose disposal in hyperglycemic, adultonset diabetics. It is hypothesized that endogenous PGE synthesis may play a role in defective insulin secretion and glucose intolerance in diabetes mellitus.